Summary The health watchdog NICE recommends that patients with moderate‐to‐severe psoriasis should try at least two non‐biologic systemic therapies before they are eligible for biologic drugs. Unfortunately, there are few studies which compare different drugs and give information about how long they are effective. The authors, based in several major centres in the UK, performed a systematic review (a type of high quality way of reviewing the scientific research) of the major non‐biologic drugs used in psoriasis, using databases including MEDLINE, PubMed and Embase. They identified 411 studies, of which 8 had recruited more than 100 patients who were included in the study, representing 4624 patients. Discontinuation because of adverse events (unwanted side effects) was commoner with treatments called fumaric acid esters (43‐46%) than with methotrexate (22%). In one study, 50% remained on methotrexate, 42% on acitretin but only 22% on ciclosporin after one year; however, this may reflect the fact that ciclosporin is generally used in short courses because of increased toxicity in long term use. In other studies, the mean time for discontinuing methotrexate ranged from 7.7 to 22.3 months, whereas for fumaric acid esters it was 28 to 50 months. 76% of patients on fumaric acid esters and 59% on methotrexate achieved a reduction in Psoriasis Area and Severity Index (PASI score) of greater than 75%. PASI is used to record the redness, thickness and scaling of a patient's psoriasis and to measure how well a treatment works, as a reduction in the PASI score means a reduction in these symptoms; for example PASI 75 means the patient has a 75% or more reduction in their PASI score from the start. The authors concluded that it was difficult to compare the available studies because they used different methods of analysis, and stressed the need for long term observational studies (i.e. where researchers observe the effect of a treatment without trying to change who is or isn't exposed to it) of good quality. Linked Article: Mason et al. Br J Dermatol 2019; 181:256–264